[SERVER-37549] Compute atClusterTime for sharded transactions using lastApplied opTime timestamps of targeted shards Created: 10/Oct/18 Updated: 06/Dec/22 Resolved: 17/Jan/19 |
|
| Status: | Closed |
| Project: | Core Server |
| Component/s: | Sharding |
| Affects Version/s: | None |
| Fix Version/s: | None |
| Type: | Task | Priority: | Major - P3 |
| Reporter: | Jack Mulrow | Assignee: | [DO NOT USE] Backlog - Sharding Team |
| Resolution: | Duplicate | Votes: | 0 |
| Labels: | None | ||
| Remaining Estimate: | Not Specified | ||
| Time Spent: | Not Specified | ||
| Original Estimate: | Not Specified | ||
| Issue Links: |
|
||||||||||||||||||||||||||||
| Assigned Teams: |
Sharding
|
||||||||||||||||||||||||||||
| Participants: | |||||||||||||||||||||||||||||
| Description |
|
The atClusterTime selection algorithm designed for the Global Point in Time reads project uses the last majority committed opTime timestamps from each targeted shard to select an atClusterTime timestamp. This is incompatible with the speculative snapshot optimization and can lead to problems on shards with enableMajorityReadConcern=false, so instead the algorithm should use the last applied opTime timestamps of the targeted shards. |
| Comments |
| Comment by Kaloian Manassiev [ 17/Jan/19 ] |
|
Folded into |